Defining Wildfire Susceptibility Maps in Italy for Understanding Seasonal Wildfire Regimes at the National Level

Abstract

Wildfires constitute an extremely serious social and environmental issue in the Mediterranean region, with impacts on human lives, infrastructures and ecosystems. It is therefore important to produce susceptibility maps for wildfire management. The wildfire susceptibility is defined as a static probability of experiencing wildfire in a certain area, depending on the intrinsic characteristics of the territory. In this work, a machine learning model based on the Random Forest Classifier algorithm is employed to obtain national scale susceptibility maps for Italy at a 500 m spatial resolution. In particular, two maps are produced, one for each specific wildfire season, the winter and the summer one. Developing such analysis at the national scale allows for having a deep understanding on the wildfire regimes furnishing a tool for wildfire risk management. The selected machine learning model is capable of associating a data-set of geographic, climatic, and anthropic information to the synoptic past burned area. The model is then used to classify each pixel of the study area, producing the susceptibility map. Several stages of validation are proposed, with the analysis of ground retrieved wildfire databases and with recent wildfire events obtained through remote sensing techniques.

Spatial wildfire hazard patterns in the Eastern Mediterranean: perspectives from a harmonised approach

“…Interesting insights are also given by the ranking of the predisposing factors in Table 5: the importance of vegetation continuity when compared to single pixel vegetation is an evident result of the analysis, with climatic and topographic classes which cannot be neglected for a good classification of the study area. This is in line with the more consolidated results from the recent work at the national scale in Italy (Trucchia et al, 2022). 2, taking as inputs the susceptibility classes (Figure 2) and the fire intensity category (Table 1)…”

“…The proposed methodology for susceptibility mapping is based on a ML model (Trucchia et al, 2022), structured as a classification task. It uses a Random Forest Classifier (RF) as an algorithm, to find a functional relation between the dependent variable (the label, that is, wildfire occurrences) and the independent variables (that is, the predisposing factors).…”

“…The Authors with the presented work continue a research framework started at the local level (Tonini et al, 2020) and recently expanded at the national scale (Trucchia et al, 2022). The described work is one of the first attempts (up to the Authors' knowledge), to model wildfire hazard at the supernational scale, where useful information can be drawn in view of cross-border wildfire management, supporting also the European Civil Protection Mechanism.…”

Wildfire hazard mapping in the eastern Mediterranean landscape

“…In this work, we describe in detail a possible methodology for wildfire hazard assessment at supranational level, along the lines of a research framework started at local level for the Liguria Region in Italy (Tonini et al 2020) and recently expanded at national scale (Trucchia et al 2022). Such methodology is based on the drafting of a contingency hazard matrix aiming at coupling the information related to wildfire susceptibility and potential fireline intensity.…”

Comparison of the Machine Learning Methods to Predict Wildfire Areas

“…Each pixel of the study are is classifed by the model. Experimental results show the ability of RF to notice the most senstive areas with defined factors [17].…”